Introduction: Diabetic retinopathy (DR) is neglected in the planning of health services in Brazil. The purpose of this study was to investigate the frequency of Diabetic Retinopathy in patients with type 2 diabetes mellitus (DM2) registered in the Unified Health System (SUS) of the semi-arid of Paraíba (PB-Brazil), using a score developed by the researchers to track this condition. Methods: A cross-sectional study was carried out, in which DM2 patients registered at SUS in the city of S?o Mamede-PB was recruited. Ophthalmological and clinical data were collected, including: body mass index (BMI), waist circumference (WC), waist circumference/height ratio (WC/stature), smoking history, physical activity, time of diabetes and its control, blood pressure values and comorbidities. With these data, an original score for DR screening (DRSS) was performed. Qualitative variables were expressed by absolute and relative frequencies, and quantitative variables were expressed by means and standard deviations. p value ≤ 0.05 was used to reject the null hypothesis. Results: Ninety seven (64.7%) type 2 diabetic patients were recruited. Participants were predominantly women (64.9%), with a mean age of 65.8 years, of short stature (mean of 1.56 m), high BMI, with a predominance of the WC/stature ratio equal to or greater than 0.60, smokers and patients with systemic arterial hypertension (SAH). The main ophthalmological findings were: vascular changes compatible with Grade I/II Hypertensive Retinopathy (72.1%) and cataracts in about 50% of the eyes. The prevalence of DR was 12.4%. According to the DRSS results, it was found that most patients (85.6%) had a high risk score equal to or greater than 60 of developing DR. It was observed that the variables: WC, physical activity, time of diabetes and SAH showed a significant association with the risk of developing DR. In addition, BMI and WC/height ratio with higher values had a high risk of developing DR. Conclusion: There was association between medium and high DRSS risk diabetics with the probability of developing DR, recommending that all people with these characteristics should be referred to the specialist in order to screen for DR and other morbidities caused by diabetes.